Trend with pinholes and resistance?
In these trials, I used TER to look at the resistance between membranes with different amounts of pinholes. Finding a trend would produce a quicker way to scan for pinholes as opposed to checking each slot on the wafer with the microscope. The table below is my raw data.
| Wafer | Number of Pinholes | Resistance (ohms) | Average (ohms) |
| 1 | 585
571 |
578 | |
| TEM23 | 1 | 579
580 |
579.5 |
| TEM23 | 1 | 602
BROKE |
602 |
| SC024 | 1 | 660
672 |
666 |
| SC075 | 1 | 641
631 |
636 |
| TEM23 | 2 | 450
558 |
504 |
| TEM23 | 2 | 616
611 |
613.5 |
| TEM23 | 2 | 617
619 |
618 |
| SC084 | 2 | 630
606 |
618 |
| SC084 | 2 | 558
545 |
551.5 |
| SC084 | 2 | 628
629 |
628.5 |
| TEM23 | 3 | 617
610 |
613.5 |
| TEM23 | 3 | 627
591 |
609 |
| ~50 | 1070
1087 |
1078.5 | |
| BROKEN | 700
689 |
694.5 |
I could not find a trend, and I was extremely surprised when seeing a membrane with a high pinhole density (~50 pinholes) having a high resistance. I expected the resistance to be lower than the resistance in the membranes with very few pinholes.
Inconsistency in the data could be due to the construction of the sepcon. For instance, the top electrode of the EVOM needs be held each time a measurement is being taken as opposed to just being latched on when using a commercial device; the sepcon does not sit well in the device.
In conclusion, I have found no differences in resistance between 1, 2, and 3 pinholes in a membrane. Is anyone seeing something I’m not?